comparison to a recombinant human IFN-a standard also included in the assay (EMD

Biosciences, San Diego, CA; 3.84 x 108 IU/mg).

Statistical Analysis

 Percentage data were transformed by arcsin transformation before analysis. Probability

values for percentage data are based on analysis of arcsin-transformed data while least-squares

means are from analysis of untransformed data.

 The proportion of oocytes that cleaved, that developed to the blastocyst stage on day 7

(experiment 2) or day 8 (experiment 1) and that developed to advanced blastocyst stages

(expanded, hatching or hatched) on day 7 (experiment 2) or day 8 (experiment 1) were calculated

for each replicate in each experiment. Treatment effects were analyzed using least-squares

analysis of variance using the General Linear Models procedure of SAS (SAS for Windows,

version 9.0, SAS Inst., Inc., Cary, NC). The model included the main effects of replicate and

treatment. All values reported are least-squares means SEM.

 Recovery rate in experiment 1, as well as embryo length and IFN-T secretion in both

experiments, were analyzed by analysis of variance using the GLM procedure of SAS. The

statistical model in experiment 1 included treatment, flush type (i.e. slaughter vs. live animal),

cow(flush type x treatment) and treatment x flush type. For experiment 2, the statistical model

included replicate, treatment, lactation and all two-way interactions. For IFN-T secretion, data

were analyzed with and without embryo length as a covariate. All values obtained from the

GLM procedure are reported as least-squares means SEM. The correlation between embryo

length and IFN-T secretion was analyzed using the CORR procedure of SAS.

 Embryo recovery in experiment 2 and the proportion of embryos that had a visible

embryonic disc at day 14 after ovulation in experiment 1 were analyzed by logistic regression